Dosimetric results in treatments of neuroblastoma and neuroendocrine tumors with (131)I-metaiodobenzylguanidine with implications for the activity to administer.

LUND UNIVERSITY PO Box 117 221 00 Lund +46 46-222 00 00

Dosimetric results in treatments of neuroblastoma and neuroendocrine tumors with (131)I-metaiodobenzylguanidine with implications for the activity to administer.

MINGUEZ GABINA, PABLO; Flux, Glenn; Genolla, Jose; Guayambuco, Sonia; Delgado, Alejandro; Fombellida, Jose Cruz; Sjögreen Gleisner, Katarina

Published in:

Medical Physics

DOI:

10.1118/1.4921807

2015

Document Version:

Peer reviewed version (aka post-print) Link to publication

Citation for published version (APA):

MINGUEZ GABINA, PABLO., Flux, G., Genolla, J., Guayambuco, S., Delgado, A., Fombellida, J. C., & Sjögreen Gleisner, K. (2015). Dosimetric results in treatments of neuroblastoma and neuroendocrine tumors with (131)I- metaiodobenzylguanidine with implications for the activity to administer. Medical Physics, 42(7), 3969-3978.

https://doi.org/10.1118/1.4921807 Total number of authors:

7

Creative Commons License:

CC BY-NC-ND

General rights

Unless other specific re-use rights are stated the following general rights apply:

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research.

• You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

Read more about Creative commons licenses: https://creativecommons.org/licenses/

Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Download date: 06. Nov. 2022

1 Dosimetric results in treatments of neuroblastoma and neuroendocrine tumors with ¹³¹I- metaiodobenzylguanidine with implications for the activity to administer

Pablo Mínguez^a)

Department of M edical Radiation Physics, Lund University, Lund, Sweden

Department of M edical Physics, Gurutzeta/Cruces University Hospital, Barakaldo, Spain

5

Glenn Flux

Joint Department of Physics, Royal M arsden NHS Foundation Trust & Institute of Cancer Research, Sutton, UK

José Genollá

Department of Nuclear M edicine, Gurutzeta/Cruces University Hospital, Barakaldo, Spain

10

S onía Guayambuco

Department of Nuclear M edicine, Gurutzeta/Cruces University Hospital, Barakaldo, Spain

Alejandro Delgado

Department of Nuclear M edicine, Gurutzeta/Cruces University Hospital, Barakaldo, Spain José Cruz Fombellida

15

Department of Nuclear M edicine, Gurutzeta/Cruces University Hospital, Barakaldo, Spain

Katarina S jögreen Gleisner

Department of M edical Radiation Physics, Lund University, Lund, Sweden

20

Purpose: The aim was to investigate whole-body and red-marrow absorbed doses in treatments of neuroblastoma (NB) and adult neuroendocrine tumors (NET) with ¹³¹I- metaiodobenzylguanidine (mIBG), and to propose a simple method for determining the activity to administer when dosimetric data for the individual patient are not available.

25

Methods: Nine NB patients and six NET patients were included, giving in total 19 treatments as four patients were treated twice. Whole-body absorbed doses were determined from dose- rate measurements and planar gamma-camera imaging. For six NB and five NET treatments, red-marrow absorbed doses were also determined using the blood-based method.

Results: Dosimetric data from repeated administrations in the same patient were consistent.

30

In groups of NB and NET patients, similar whole-body residence times were obtained, implying that whole-body absorbed dose per unit of administered activity could be reasonably well described as a power function of the patient mass. For NB, this functional form was found to be consistent with dosimetric data from previously published studies. The whole-

2 body to red-marrow absorbed dose ratio was similar among patients, with values of 1.4±0.6 to 35

1.7±0.7 (1 standard deviation) in NB treatments, and between 1.5±0.6 and 1.7±0.7 (1 standard deviation) in NET treatments.

Conclusions: The consistency of dosimetric results between administrations for the same patient supports prescription of the activity based on dosimetry performed in pre-treatment studies, or during the first administration in a fractionated schedule. The expressions obtained 40

for whole-body absorbed doses per unit of administered activity as a function of patient mass for NB and NET treatments are believed to be a useful tool to estimate the activity to administer at the stage when the individual patient biokinetics has not yet been measured.

Key words: Neuroblastoma, Neuroendocrine tumors, Treatment schedule, ¹³¹I-mIBG.

45

1. INTRODUCTION

In treatments of neuroblastoma (NB) and adult neuroendocrine tumors (NET), surgery is the first-line therapy with the aim of achieving a complete cure^1,2. If radical surgery is not 50

feasible, multimodal treatment options include surgery, chemotherapy, external beam radiotherapy (EBRT) and molecular radiotherapy (MRT)³. ¹³¹I-mIBG therapy is administered for inoperable pheocromocytomas, paragangliomas, carcinoid tumors, stage III or IV relapsed or primary refractory NBs and metastatic or recurrent medullary thyroid cancers⁴. The number of patients treated with ¹³¹I-mIBG is usually limited compared to the total number of NB and 55

NET patients as ¹³¹I-mIBG therapy is frequently considered only when other treatment modalities have been exhausted.

During the last two decades several studies have been published on the use of ¹³¹I-mIBG for NB and NET treatments^5-12, and guidelines have been provided by the European Association of Nuclear Medicine (EANM)⁴. The administered activity is most commonly prescribed using 60

3 fixed activities^13,14 or a specified activity per patient mass^15-18 or per body surface¹⁹. Some studies have shown an improved treatment outcome with increased administered activities^20,21. In NB treatments, the most established schedule is given by Gaze et al.¹⁷. In this protocol, two treatment administrations separated by a fortnight are given where the first is prescribed as activity per body mass (444 MBq/kg), and the second is tailored to deliver a 65

whole-body absorbed dose of 4 Gy in total for the two administrations. In NET treatments, where dosimetric data are still limited²², established schedules are currently lacking, although in principle a similar dosimetry-based approach could be adopted. In both NB and NET treatments, there is a need to compile experience and working knowledge of clinically obtained whole-body absorbed doses per unit of administered activity. Such information can 70

be used to improve estimates of the activity to administer at the stage when dosimetric data are lacking, such as for the first activity administration in a fractionated schedule.

Hematologic toxicity is dose limiting in ¹³¹I-mIBG therapy^23-25. When activities above 444 MBq/kg are administered, harvesting of autologous tumor-free, hematopoietic stem cells must be performed before treatment²⁶. In the schedule by Gaze et al.¹⁷, stem-cell rescue is 75

performed after approximately four weeks from the first administration, when the activity in the body has decreased below 30 MBq. When stem cells are not available, it must be ensured that the administered activity does not exceed levels that may induce non-tolerable red- marrow absorbed doses. Here, the use of whole-body dosimetry as a surrogate for red-marrow dosimetry has been established²⁷. This obviates the need for repeated blood sampling, which 80

is considered invasive, particularly in children. However, as pointed out in a recent review¹², no study has yet focused on the difference between whole-body and red-marrow absorbed doses.

This study reports on dosimetric results from ¹³¹I-mIBG NB and NET treatments in the Gurutzeta-Cruces University Hospital during the last six years. The aim is to investigate 85

4 absorbed doses for whole body and red marrow. A further aim is to propose a simple method for determining the activity to administer at the stage when dosimetric data are not available for the individual patient, based on data acquired in this study and in the context of other published studies. The recommendations of the EANM²⁸ have been taken into account in the composition of the paper.

90

2. METHODS

2.A. Patient population and administrations NB treatments

Nine patients (six male and three female, age 3y−22y) with relapsed stage-4 NB were 95

included. Three patients were given two treatments separated by more than one year, with the result that twelve treatments were considered in total. Further on, NB treatments are denoted TNB1―TNB9, with postscripts a and b to indicate repeated treatment of the same patient. A summary of treatment data including patient mass, mp, administered activity, Aadm, and performed measurements is given in Table 1. Treatments TNB1a, TNB2, TNB3, TNB4, TNB6a, 100

TNB7 and TNB8 were performed following the schedule by Gaze et al.¹⁷. Treatments TNB1b, TNB5, TNB6b, TNB4 and TNB9b were performed without concomitant chemotherapy and stem cell support in one or two fractions (see Table 1) aiming at giving a whole-body absorbed dose, Dwb, of 2 Gy. Treatment TNB9a was performed with concomitant chemotherapy and stem cell support but did not follow the schedule by Gaze et al.¹⁷ to avoid exceeding Aadm of 105

37 GBq. The timing of stem cell support was approximately 4 weeks post-¹³¹I-mIBG administration, as determined using dose-rate measurements.

110

5

T reatment mp

(kg)

Aadm (GBq) Performed measurements

Adm 1

Adm

2 T otal Adm 1

Adm 2

TNB1a 11 5.0 5.5 10.5 DR/BD DR

TNB1b (+13 months)* 13 5.5 N/A 5.5 DR N/A

TNB2 13 5.7 13.7 19.4 DR/BD DR

TNB3 14 6.3 6.7 13.0 DR/BD /PL DR/BD /PL

TNB4 17 7.7 6.5 14.2 DR DR

TNB5 18 4.2 3.7 7.9 DR DR

TNB6a 20 9.0 8.3 17.3 DR/BD DR

TNB6b (+35 months)* 24 10.8 N/A 10.8 DR N/A

TNB7 22 9.8 9.8 19.6 DR DR

TNB8 32 13.0 14.4 27.4 DR DR

TNB9a 63 11.1 22.2 33.3 DR/BD DR

TNB9b (+16 months)* 63 10.5 10.4 20.9 DR/BD /PL DR/BD /PL

Table 1. Data of NB treatments. DR=Dose-rate measurement. BD=Blood dosimetry. PL=Planar imaging.

Adm=Administration. *Time between repeated treatments.

NET treatments 115

Six NET patients were included (five female and one male, age 21y−80y). One patient was treated twice, so in total seven treatments were considered. Further on, NET treatments are denoted TNET1―TNET6, with postscripts a and b to indicate repeated treatment. A summary of treatment data is given in Table 2. For TNET2, TNET3, TNET5, TNET6a and TNET6b, a pre- treatment dosimetry study was also performed approximately one month before treatment.

120

Table 2. Data of NET treatments, including diagnosis. DR=Dose-rate measurement. BD=Blood dosimetry.

PL=Planar imaging. Pre-Tr Adm=Pre-treatment administration. Adm=Administration. *Time between repeated treatments.

125

T reatment mp

(kg)

Aadm (GBq) Performed measurements

Pre-Tr Adm

Adm 1

Adm 2

T otal (Adm 1 +

Adm 2)

Pre-Tr Adm

Adm 1

Adm 2

TNET1 (paraganglioma) 58 N/A 16.2 N/A 16.2 N/A DR N/A

TNET2 (carcinoid tumor) 44 0.44 21.2 N/A 21.2 PL/BD DR N/A TNET3 (pheochromocytoma) 66 0.071 14.4 N/A 14.4 PL/BD DR N/A

TNET4 (pheochromocytoma) 49 N/A 5.6 7.8 13.4 N/A DR DR

TNET5 (carcinoid tumor) 80 0.23 9.0 8.6 17.6 PL/BD DR/BD DR TNET6a (pheochromocytoma ) 49 0.19 7.8 N/A 7.8 PL/BD DR N/A TNET6b (pheochromocytoma) (+1 year)* 49 0.20 8.4 9.6 18.0 PL/BD DR DR

6 Specific activity of the administered ¹³¹I-mIBG was 1110 MBq/mg. In both NB and NET treatments, the average time as inpatients was five days in each administration (range four−six days) and patients were released according to Spanish national regulations, in agreement with recommendations of the IAEA²⁹. The therapeutic use of ¹³¹I-mIBG is approved by the Spanish Agency of Medicines and Medical Devices and informed consent from all patients, or 130

from their parents in the case of children, was obtained.

2.B. Data acquisition and activity quantification

Dose-rate measurements were performed during the time as inpatients for estimation of the whole-body time-activity curve (see Tables 1 and 2). Measurements were performed using a handheld, pressurized ion-chamber survey-meter, Inovision Model 451P, Fluke Biomedical 135

(Eindhoven, The Netherlands). Acquisitions were made at distances of 1 m and 2 m from standing patients at marked positions on the floor, in both anterior and posterior directions.

The height of the detector in relation to the floor was held constant with reference to an external mark, and all measurements were made by trained staff. The first measurement was performed immediately after the administration to obtain a reading corresponding to the total 140

Aadm. Remaining measurements were made approximately every two hours during the first day, every four hours during the second day and every six hours during the remaining days, aiming at performing acquisitions after bladder voids. In total this yielded approximately 20 time points for each treatment. The signal-to-noise ratio, estimated by dividing the patient readings with the variability in background readings, was above 20 in all measurements, and 145

dead-time effects were negligible. A sequence of whole-body time-activity values, Awb(t), was determined from the anterior and posterior readings, RA(t) and RP(t), for each patient-detector distance according to

𝐴_𝑤𝑏(𝑡) = 𝐴_𝑎𝑑𝑚 √𝑅_𝐴(𝑡)𝑅_𝑃(𝑡)

√𝑅_𝐴(0)𝑅_𝑃(0)= 𝐴_𝑎𝑑𝑚 𝑟(𝑡)

(1)

7 where r(t) were the relative values obtained from measurement, with r(0) =1. The Awb(t) values from measurements at 1 m and 2 m differed less than 5%, and the average value was 150

therefore used. A dose calibrator, Capintec CRC®-15R, (Capintec, Inc Ramsey, NJ, USA), was used for measurements of Aadm.

Planar imaging using a gamma camera was employed to estimate the whole-body time- activity curve in NET pre-treatment dosimetric studies, and also in two NB treatments for comparison to dose-rate meter derived values (Tables 1 and 2). Acquisitions were made 155

employing a dual-head General Electric (GE, Fairfield, CT, USA) Infinia Hawkeye gamma camera, with a crystal thickness of 9.5 mm and equipped with High-Energy General-Purpose collimators. A scan speed of 12 cm/min, a matrix size of 2561024, and an energy window of 20% centered at 364 keV were used. For the NET pre-treatment dosimetric studies, in which acquisitions were performed a few minutes, 24 h, 48 h and 120 h after the administration, the 160

administered activity was low and dead-time effects were thus negligible. To avoid dead-time effects for NB patients, where pre-treatment imaging was not performed, the first therapy administration was separated into two fractions. Approximately 370 MBq was injected and a whole-body scan was performed. Immediately after this acquisition, the rest of the activity was injected. The remaining acquisitions were performed approximately at 48 h and 115 h. In 165

all the acquisitions performed, the count rate was below 10000 counts per second. The whole- body activity was determined using Eq. (1), where RA(t) and RP(t) were then the net count rates in regions of interest (ROIs) encompassing the body in anterior and posterior images, subtracted by the count rate in background ROIs rescaled to the area of the whole-body ROI, to partly compensate for septal penetration and scatter.

170

Blood sampling was performed in six NB treatments, five NET pre-treatment studies, and one NET treatment, for the purpose of red-marrow dosimetry (see Tables 1 and 2). For NB patients blood sampling was performed at a few minutes, 6 h, 24 h, 48 h, 72 h, and 96 h or

8 115 h after injection, whereas for NET patients, blood sampling was made at a few minutes, 6 h, 24 h, 48 h, and 120 h. Blood samples of 1 ml or 2 ml volume were prepared using a pipette, 175

and were then allowed to decay to avoid dead-time effects. Measurements were performed using a calibrated γ-well counter 1282 Compugamma CS LKB Wallac (Melbourne, Australia). The activity concentration was determined by dividing the obtained count rate by a pre-determined calibration factor and the sample volume.

2.C. Dosimetric calculations 180

Dwb were calculated following the standard MIRD methodology³⁰, as described in Appendix A. Blood-based calculation of red-marrow absorbed dose, Drm, was carried out according to procedures in the EANM guidelines²⁶ and to the MIRD formalism, including source terms from activity in the red marrow and in the remainder of the body for the self- and cross- absorbed dose, respectively, as described in Appendix B. Additionally, the method described 185

by Traino et al.³¹ was used.

The standard deviations in Dwb and Drm were estimated by uncertainty propagation³², as described in Appendix C. For both Dwb and the whole-body residence time, τwb, the relative standard deviation was estimated to be 20%. The main contribution to Drm was the cross- absorbed dose from the remainder of the body, and the relative standard deviation for Drm

190

was thus also estimated to be 20%. These uncertainties are in line with values suggested by others^33,34.

2.D. Hematologic toxicities

The post-therapy platelet, neutrophil and leukocyte nadir was obtained in NET treatments in order to study hematologic toxicity. The grade of toxicity was analyzed according to the 195

Common Terminology Criteria of Adverse Events (CTCAE), version 3³⁵ (available at http://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/ctcaev3.pdf). In

9 NB treatments the grade of hematologic toxicity was not quantified, since in the majority of treatments the intent was aplasia.

200

3. RESULTS 3.A. NB treatments

Table 3 summarizes values obtained for τwb and Dwb/Aadm. The mean value of Dwb/Aadm was 0.22 ±0.04 (1 standard deviation) Gy/GBq (median 0.22 Gy/GBq). For all treatments except 205

TNB2, where the patient suffered from nephropathy and thus had a shorter τwb, values of τwb

were within 25.1 h - 29.3 h (mean 27.1 ±5.4 h, median 26.7 h). For the patients that followed the schedule by Gaze et al.¹⁷, that is, those who were prescribed a Dwb of 4 Gy in two administrations, the prescription was followed to within 0.1 Gy.

T reatment τwb (h) Dwb/Aadm (Gy/GBq) Dwb (Gy)

Adm 1 Adm 2 Adm 1 Adm 2 T otal Adm 1 Adm 2 T otal

TNB1a 26.5±5.3 25.5±5.1 0.38±0.08 0.36±0.07 0.37±0.07 1.9±0.4 2.0±0.4 3.9±0.8

TNB1b 25.1±5.0 N/A 0.33±0.07 N/A 0.33±0.07 1.8±0.4 N/A 1.8±0.4

TNB2 15.9±3.2 16.0±3.2 0.21±0.04 0.20±0.04 0.21±0.04 1.2±0.2 2.8±0.6 4.0±0.8 TNB3 26.7±5.3 28.2±5.6 0.30±0.06 0.33±0.07 0.32±0.06 1.9±0.4 2.2±0.4 4.1±0.8 TNB4 28.6±5.7 28.3±5.7 0.29±0.06 0.28±0.06 0.28±0.06 2.2±0.4 1.8±0.4 4.0±0.8 TNB5 25.5±5.1 25.2±5.0 0.24±0.05 0.22±0.04 0.23±0.05 1.0±0.2 0.8±0.2 1.8±0.4 TNB6a 27.3±5.5 27.9±5.6 0.23±0.05 0.24±0.05 0.24±0.05 2.1±0.4 2.0±0.4 4.1±0.8

TNB6b 26.3±5.3 N/A 0.19±0.04 N/A 0.19±0.04 2.0±0.4 N/A 2.0±0.4

TNB7 26.0±5.2 26.2±5.2 0.19±0.04 0.20±0.04 0.20±0.04 1.9±0.4 2.0±0.4 3.9±0.8 TNB8 26.6±5.3 26.1±5.2 0.15±0.03 0.15±0.03 0.15±0.03 1.9±0.4 2.1±0.4 4.0±0.8 TNB9a 29.3±5.9 28.9±5.8 0.09±0.02 0.09±0.02 0.09±0.02 1.0±0.2 1.9±0.4 2.9±0.6 TNB9b 28.8±5.8 28.9±5.8 0.09±0.02 0.09±0.02 0.09±0.02 0.9±0.2 0.9±0.2 1.8±0.4

Table 3. Results for τwb, Dwb/Aadm and Dwb in NB treatments

210

Data in Table 3 are based on probe-based dose-rate measurements. Figure 1 shows a comparison between Awb (t) derived from dose-rate measurements and gamma camera images for TNB3. Differences between values obtained from imaging and dose-rate measurements were within 10% for both TNB3 and TNB9b.

215

10

0 1000 2000 3000 4000 5000 6000 7000

0 20 40 60 80 100 120

Awb (MBq) (Dose rate method) Awb (MBq) (Planar imaging method)

A wb (MBq)

t (h)

0 1000 2000 3000 4000 5000 6000 7000

0 20 40 60 80 100 120

Awb (MBq) (Dose rate method) Awb (MBq) (Planar imaging method)

t (h) A wb (MBq)

Figure 1. Results for Awb(t) , for the first (left panel), and second administration (right panel), in TNB3. Open symbols are from dose-rate measurements, and closed symbols are from whole-body planar images acquired at

48h and 115h.

220

Table 4 shows the results of Drm and Drm/Aadm. For TNB1b, TNB4, TNB5, TNB6b, TNB7 and TNB8 red-marrow dosimetry could not be performed due to lack of blood samples. Following the procedure by Traino et al.³¹ values obtained for Drm were lower, but within 10 % of values of Drm in Table 4. The self-absorbed dose was between 9% and 11% of the total Drm. For the two 225

treatments where measurements were made in both administrations (TNB3 and TNB9b), the ratio Drm/Aadm was approximately equal. For the other treatments, Drm for the second administration was thus estimated by assuming an equal Drm/Aadm between administrations.

Table 4 also shows the ratio Dwb/Drm, which ranged from 1.4 to 1.7 (mean 1.5±0.6, median 1.5).

230

T reatment

Adm 1 Adm 2 Both Adm

Drm

(Gy)

Drm/Aadm

(Gy/GBq) Drm

(Gy)

Drm/Aadm

(Gy/GBq) Drm

(Gy) Dwb / Drm

TNB1a 1.4±0.3 0.27±0.05 1.5±0.3* N/A 2.9±0.6 1.4±0.6 TNB2 0.8±0.2 0.15±0.03 2.0±0.4* N/A 2.9±0.6 1.4±0.6 TNB3 1.1±0.2 0.18±0.04 1.3±0.3 0.19±0.04 2.4±0.5 1.7±0.7 TNB6a 1.0±0.3 0.14±0.03 1.2±0.2* N/A 2.5±0.5 1.6±0.6 TNB9a 0.6±0.1 0.06±0.01 1.3±0.3* N/A 1.9±0.4 1.5±0.6 TNB9b 0.6±0.1 0.05±0.01 0.6±0.1 0.06±0.01 1.2±0.2 1.5±0.6

Table 4. Results for Drm and Drm/Aadm, in administrations 1 and 2, in NB treatments. For both administrations, the values of the total Drm and Dwb/Drm are shown. * Extrapolated values obtained by assuming the value of Drm/Aadm obtained for the first

administration.

11 235

3.B. NET treatments

Table 5 summarizes the values obtained for τwb and Dwb/Aadm. The mean value of Dwb/Aadm

was 0.16±0.03 Gy/GBq (median 0.13 Gy/GBq). With the exception of TNET4 and TNET5, who 240

suffered from a large tumor burden and thus had notably longer τwb, values of τwb were within the range 31.0 h ― 35.2 h (mean 33.1 ±6.6 h, median 33.3 h).

T reatment τwb (h) Dwb/Aadm (Gy/GBq) Dwb (Gy)

Adm 1 Adm 2 Adm 1 Adm 2 T otal Adm 1 Adm 2 T otal

TNET1 32.1±6.4 N/A 0.10±0.02 N/A 0.10±0.02 1.7±0.3 N/A 1.7±0.3 TNET2 35.2±7.0 N/A 0.15±0.03 N/A 0.15±0.03 3.1±0.6 N/A 3.1±0.6 TNET3 31.0±6.2 N/A 0.09±0.02 N/A 0.09±0.02 1.3±0.3 N/A 1.3±0.3 TNET4 82.2±16.4 82.6±16.5 0.30±0.06 0.31±0.06 0.31±0.06 1.7±0.3 2.4±0.5 4.1±0.8 TNET5 69.5±13.9 58.2±11.6 0.17±0.03 0.14±0.03 0.15±0.03 1.5±0.3 1.2±0.2 2.7±0.5 TNET6a 33.1±6.6 N/A 0.13±0.03 N/A 0.13±0.03 1.0±0.2 N/A 1.0±0.2 TNET6b 33.7±6.7 33.5±6.7 0.13±0.03 0.14±0.03 0.13±0.03 1.1±0.2 1.3±0.3 2.4±0.5

Table 5. Results for τwb, Dwb/Aadm and Dwb in NET treatments.

245

Table 6 shows Drm/Aadm obtained for pre-treatment dosimetric studies. In TNET1 and TNET4 blood samples could not be obtained. Following the procedure by Traino et al.³¹ lower values for Drm were obtained, but within 10% of values of Drm in Table 6. The self-absorbed dose was between 14% and 18% of the total Drm. For TNET5, Drm was also calculated during the first treatment administration. The value of Drm/Aadm obtained was 0.10±0.02 Gy/GBq, which 250

was thus in agreement with that of the pre-treatment study. Table 6 also shows the ratio Dwb/Drm, which ranged between 1.5 and 1.7 (mean 1.6±0.6, median 1.7).

255

12

Pre-treatment study

Drm/Aadm

(Gy/GBq) Dwb / Drm

TNET2 0.08±0.02 1.7±0.7 TNET3 0.06±0.01 1.5±0.6 TNET5 0.09±0.02 1.7±0.7 TNET6a 0.07±0.01 1.7±0.7 TNET6b 0.08±0.02 1.6±0.6

Table 6. Results for Drm/Aadm, and Dwb/Drm in pre-treatment dosimetric studies in NET patients.

260

The grade of toxicity in NET treatments is shown in Table 7. No correlation between the grade of toxicity in platelets, leukocytes and neutrophils and Dwb or Drm was found. However, there was a tendency that toxicity was more pronounced for elderly patients than for younger ones.

265

T reatment

Grade of toxicity Patients’ age

(y) Platelets Leukocytes Neutrophils

TNET1 None None None 24

TNET2 3 4 4 73

TNET3 4 4 4 62

TNET4 3 3 2 60

TNET5 3 3 2 80

TNET6a None None None 27

TNET6b None None None 28

Table 7. Grade of toxicity in NET treatments.

3.C. Analysis of dosimetric results for NBs and NETs

Figure 2 shows the ratio Dwb/Aadm for both NB and NET treatments, and a large variability between patients can be observed. Values of Dwb/Aadm for repeated administrations (including 270

treatments and pre-treatment dosimetric studies) in the same patient were consistent, thus supporting the concept of performing absorbed dose planning for subsequent administrations.

13

0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 0,4

Pre-treatment

Treatment administration 1 Treatment administration 2

TNB1 TNB2 TNB3 TNB4 TNB5 TNB6 TNB7 TNB8 TNB9 TNB10 TNB11 TNB12 TNET1 TNET2 TNET3 TNET4 TNET5 TNET6 TNET7

D wb / A adm(Gy/GBq)

Figure 2. Results obtained for Dwb/Aadm in NB treatments and in NET pre-treatment and treatment studies.

As seen in Table 3, a similar τwb was obtained in NB treatments, except for TNB2. If inserting 275

one single value of τwb into Eq. (A1), the ratio Dwb/Aadm follows a power dependence with mp. Using the mean value of τwb obtained in this study, the following Equation was obtained:

( 𝐷_𝑤𝑏⁄𝐴_𝑎𝑑𝑚)_𝑁𝐵 = 3.63 𝑚_𝑝^−0.921 (2)

in unit of Gy/GBq. Results of Eq. (2) were compared to results from previously published studies, using data of Dwb/Aadm for NB treatments as retrieved from the study by Toporski et 280

al.¹¹ performed at Lund University Hospital, Sweden, and from the study by Buckley et al.²⁵ performed at Royal Marsden Hospital, UK. To rule out inconsistencies in calculation methods among centers, a small comparison exercise was undertaken sharing three sets of acquired time-dose rate data. Dwb were calculated at each of the three centers, with results obtained of within 10% from the mean value, thus supporting a combined data analysis. Figure 3 shows 285

Eq. (2) in relation to data acquired in this study, and combined with data from Toporski et al.¹¹ and Buckley et al.²⁵.

14

0 0,2 0,4 0,6 0,8 1

0 20 40 60 80 100

Minguez et al. (NB) Toporski et al.

Buckley et al.

Equation 2

D wb / A adm (Gy/GBq)

mp (kg)

Figure 3. Results for Dwb/Aadm as function of mp from Toporski et al.¹¹, Buckley et al.²⁵, and data obtained in this study. The

290

solid line shows the curve obtained from Eq. (2).

Regarding NET (Table 5), similar values of τwb were obtained for the five treatments with modest tumor burden. Inserting the mean value obtained into Eq. (A1), the following Equation was obtained:

295

( 𝐷_𝑤𝑏⁄𝐴_𝑎𝑑𝑚)_𝑁𝐸𝑇 = 4.44 𝑚_𝑝^−0.921 (3)

Figure 4 shows Dwb/Aadm as a function of mp using Eq. (3). Comparing the graphs for NB and NET treatments (Figure 4, right panel), differences in Dwb/Aadm values were small for adult masses but increased slightly for pediatric masses.

0 0,2 0,4 0,6 0,8 1

0 20 40 60 80 100

NET treatment results Equation 3

D wb / A adm (Gy/GBq)

mp (kg)

0 0,2 0,4 0,6 0,8 1

0 20 40 60 80 100

Equation 2 Equation 3

D wb / A adm (Gy/GBq)

mp (kg)

300

Figure 4. Results for Dwb/Aadm as a function of mp in NET patients (left), and (right) comparison of Eqs. (2) and (3), representing Dwb/Aadm as a function of mp for NB and NET patients, respectively.

305

15 4. DISCUSSION

In NB and NET treatments with ¹³¹I-mIBG, very different schedules have been reported^13-19, 310

often using prescriptions in terms of a fixed activity or a predetermined activity per body mass. For treatment of NB, the most widely used dose scheduling approach is that of Gaze et al.¹⁷ and is based on planning of Dwb using two treatment fractions. In NET treatments, such established schedules are still lacking, although a similar approach could be adopted to tailor Dwb with regard to the risk of inducing hematologic and non-hematologic toxicities.

315

The analysis of patient data by Eqs. (2) and (3) for NB and NET treatments, respectively, was motivated by a practical need for activity prescriptions in situations when τwb for an individual patient has not yet been measured, such as for the first activity administration in a dosimetry- based schedule. Eqs. (2) and (3) should be thus regarded as an alternative to prescribing a fixed activity or activity per body mass. For comparison with the schedule given by Gaze et 320

al.¹⁷, Eq. (2) was reformulated in terms of Dwb as a function of Aadm/mp (Figure 5). Eq. (2) has the advantage of taking the decreasing values of Swbwb into account, which implies that for the heavier patients, a lower activity can be administered. For instance, for a 70 kg patient the activity to administer is decreased by approximately 12% as compared to using 444 MBq/kg¹⁷. Comparison was also made to the work by Matthay et al.²², where a linearly 325

increasing Dwb was obtained when presented as a function of Aadm/mp, with a considerably larger variability in Dwb for higher values of Aadm/mp. Their results thus agree with Eq. (2), both concerning the linear increase and the larger variability in Dwb for higher values of Aadm/mp. Due to the limited amount of data for NET patients (Figure 4), Eq. (3) should be treated with caution and regarded as preliminary. Comparison studies with results from other 330

centers would be desirable, especially for mp values outside the included range. However, in the literature, reported dosimetry values for ¹³¹I-mIBG treatment of NET patients are still few.

In this study, Dwb/Aadm values are higher and show a wider range for NB than for NET

16 treatments consistent with those reported in Sudbrock et al.⁹ and Hindorf et al.²⁷. Figures 3 and 4, based on Eqs. (2) and (3), respectively, provide an explanation for those results based 335

on the dependence of Dwb/Aadm on mp. It is important to note that Eqs. (2) and (3) are not intended as replacement for dosimetry-based schedules, since τwb of individual patients may vary to a high degree. For instance, in TNB2, the patient (mp=13kg) suffered from a nephropathy and the obtained Dwb/Aadm value was lower that the value obtained by Eq. (2), as seen in Figure 3 and Table 3. In TNET4 and TNET5 (mp=49 kg and 80 kg, respectively), patients 340

had an extensive tumor burden and their Dwb/Aadm values were notably higher than the value obtained from Eq. (3), as seen in Figure 4 (left) and Table 5. Thus, whole-body dosimetry measurements during the first administration would still be necessary in order to calculate the activity to deliver in the second administration.

0 0,5 1 1,5 2 2,5 3 3,5 4

0 100 200 300 400 500 600 700 800

mp 100 kg mp 70 kg mp 40 kg mp 10 kg 444 MBq/kg

Aadm / m_p (MBq/kg) D wb (Gy)

345

Figure 5. Representation of Eq. (2) in terms of Dwb as a function of Aadm /mp. The vertical line represents the activity of 444M Bq/kg given in the schedule by Gaze et al.¹⁷.

Dwb is generally used as a surrogate for Drm in ¹³¹I-mIBG therapy. In this study, values of Drm

were found to be between 60% and 70% of Dwb, which is reasonable considering the modest 350

contribution of the self-absorbed dose and the values for the Swb←wb/Srm←wb ratio. In patients in whom there is red-marrow and/or bone uptake, it is generally recommended to use imaging- based estimates of Drm26,36,37, as blood-based values may underestimate the real value.

However, results from the blood-based method may be a good approximation when red-

17 marrow or bone uptake is localized to small regions³⁶. For the patients in this study where 355

red-marrow dosimetry was performed, in five patients (TNB9a, TNB9b, TNET2, TNET6a and TNET6b) uptake in red marrow or bone was seen in separate SPECT-CT studies, but involved less than 5% of the total marrow volume, thus justifying the use of the blood-based dosimetry method.

A further analysis was performed by comparison to the results from Matthay et al.²², who 360

studied dose escalation of ¹³¹I-mIBG in treatment of NB with autologous stem-cell rescue. In their work, none of the 18 patients who were given activities < 555 MBq/kg required stem cell infusion, whereas two of seven patients given 555 MBq/kg and nine of 17 patients given 666 MBq/kg required stem-cell support. Using Eq. (2) and the obtained mean value of Dwb/Drm for NB of 1.5, an estimation of Drm for values of Aadm/mp of 444, 555 and 666 365

MBq/kg was made (Figure 6). Activities of 666 MBq/kg resulted in a Drm which for most mp

exceeded the tolerance dose of approximately 1.6―2 Gy²⁵. Giving 555 MBq/kg, Drm close to 2 Gy were obtained, whereas for 444 MBq/kg, Drm were well below 2 Gy. Figure 6 also shows the Dwb as estimated from Eq. (2), indicating that on average 444 MBq/kg¹⁷ results in Dwb above 2 Gy for patients above approximately 15 kg.

370

0 1 2 3 4

0 20 40 60 80 100

666 MBq/kg WB 555 MBq/kg WB 444 MBq/kg WB 4 Gy WB limit 666 MBq/kg RM 555 MBq/kg RM 444 MBq/kg RM 2 Gy RM limit

D wb and D rm(Gy)

mp (kg)

Figure 6. Representation of Dwb and Drm as function of mp for NB patients, for Aadm of 444, 555 and 666 M Bq/kg. Values have been derived from Eq. (2) and a value of Dwb/Drm of 1.5.

18 The grade of hematologic toxicity (Table 7) shows that caution must be exercised for high- 375

activity treatments. Unlike that of Buckley et al.²⁵, this study found no correlation between the grade of hematologic toxicity and Dwb. Notably, in TNET3, a Dwb of 1.3 Gy was delivered but the patient suffered grade-4 toxicity (platelets, leukocytes and neutrophils) and in TNET4, the patient with multiple bone metastases received a Dwb of 4.1 Gy but did not exceed grade-3 toxicity. As with Dwb, no correlation was found between the grade of hematologic toxicity and 380

Drm. There are several possible reasons for the lack of correlation found between Drm and Dwb

with hematologic toxicity. Among those are the low number of patients included, the different ages of patients, the diversity of NETs (pheochromocytoma, carcinoid tumor and paraganglioma), prior hematotoxic treatments, and the way the cross-absorbed dose to the red marrow is calculated in Eq. (B1). Regarding the latter point, if the activity in the remainder of 385

the body is mainly localized in tumors, then its contribution to Drm is likely to be heterogeneous with an important proportion of the red marrow receiving absorbed doses below tolerance values. In a study of hematological toxicity in EBRT performed by Petersson et al.³⁸, it was shown that the severity of toxicity correlated with the volume fraction of red marrow that was irradiated. In this study the volume distribution of Drm was not addressed, 390

and so, depending on the tumor burden, two treatments with the same value of Drm obtained from Eq. (B1) may show different hematologic toxicity. These results indicate the need to improve currently used methods for red-marrow dosimetry in MRT, taking the heterogeneous distribution of internal absorbed doses into account.

395

5. CONCLUSIONS

In treatments with ¹³¹I-mIBG, the activity to administer in order to give a prescribed Dwb

varies from patient to patient. In this study, consistent values of Dwb/Aadm were obtained when

19 determined for different administrations in the same patient, whereas a considerable variation 400

was seen among patients. These results thus support the use of absorbed-dose planning for multiple-fraction treatment. Moreover, an expression was proposed for prescription of the activity for the first administration, which takes into account the dependence of Dwb/Aadm on mp, to be used at the stage when dosimetric data for the individual patient have not yet been measured. For red marrow, Drm was found to be between 60% and 70% of Dwb.

405

Competing interests

The authors declare that they have no competing interests.

APPENDIX A: WHOLE-BODY ABSORBED DOSE (Dwb) Dwb is given by:

410

wb wb wb adm wb

wb wb

wb A S A S

D ^~ _ 



where A~_wb

is the cumulated activity in the whole body (wb) and Swb←wb is the whole-body absorbed dose per cumulated activity in wb, calculated according to Cristy et al.³⁹:

921 . 4 0

10 34 .

1 ^{ }

_wb   _p

wb m

S (A2) in Gy MBq^-1 h^-1, and mp is the patient mass (kg). The residence time τwb is determined from the values r(t) obtained from measurement (Eq. (1)), by fitting one exponential function for each of the n components and performing integration, according to:

415



 

 ⁿ

i i

i i wb

a a

1

1

  (A3)

where coefficient ai is the initial value for component i and i is the effective half-life, in unit h^-1, for the respective component²⁵. The value of n was set to 3 and 2, for dose-rate measurements and gamma-camera imaging, respectively.

420

20 APPENDIX B: RED-MARROW ABSORBED DOSE (Drm)

Drm is given by:

rb rm rb rm rm rm

rm A S A S

D ~ _ ~ _

(B1)

where A~_rm

and A~_rb A~_wb A~_rm are the cumulated activities in the red marrow (rm) and the remainder of the body (rb), respectively, Srm←rm is the factor describing the self-absorbed dose 425

from activity residing in rm, and Srm←rb is the factor describing the cross-absorbed dose from activity residing in rb and is given by the expression:

rb rm rm rm rb wb wb rm rb

rm m

S m m S m

S _  _  _ (B2)

where Srm←wb is the factor describing the cross-absorbed dose from activity residing in wb, and mwb, mrb , and mrm, are the masses of wb, rb and rm, respectively. S-values and values of 430

mwb, mrb and mrm were obtained for the male and female reference phantoms in OLINDA/EXM⁴⁰, and were scaled to the mass of the individual patient.

A~rm

was obtained following:

rm blood

rm A RMBLRm A  ]~ 

~ [

(B3)

where A]~ _blood

[ is the cumulated activity in blood per unit of volume, and RMBLR is the red 435

marrow-to-blood activity concentration ratio, which was set to 1^41,42.